賦時(shí)離散事件系統(tǒng)的故障診斷方法研究
發(fā)布時(shí)間：2019-05-22 12:51
【摘要】：離散事件系統(tǒng)是以狀態(tài)系列和事件系列刻畫系統(tǒng)演化過程的邏輯類系統(tǒng)。此類系統(tǒng)顯著的功能是利用控制策略,只允許某些符合策略的事件系列發(fā)生以達(dá)到某種工程需求;普遍應(yīng)用于通訊、交通等公共服務(wù)設(shè)施和計(jì)算機(jī)信息處理等重要領(lǐng)域。隨著科學(xué)技術(shù)的發(fā)展,此類系統(tǒng)結(jié)構(gòu)日趨復(fù)雜,故障頻發(fā);以往的故障診斷技術(shù)益發(fā)捉襟見肘。后由Sampath等人提出離散事件系統(tǒng)的基于診斷器診斷的方法解決了這一難題。基于診斷器診斷方法其實(shí)是對(duì)系統(tǒng)擴(kuò)充性地構(gòu)造其診斷器,診斷器可與系統(tǒng)并行運(yùn)行,并通過分析診斷器狀態(tài)判斷系統(tǒng)是否發(fā)生故障。盡管由Sampath等提出的離散事件系統(tǒng)的故障診斷方法能有效地診斷系統(tǒng)故障,但是并未考慮在系統(tǒng)故障診斷期間的安全性問題,為此Paoli等人在Sampath的基礎(chǔ)上進(jìn)一步研究了離散事件系統(tǒng)的安全診斷方法。此方法已被推廣至各類系統(tǒng),用于解決系統(tǒng)故障的安全性診斷問題。已推廣應(yīng)用的系統(tǒng)包括:隨機(jī)離散事件系統(tǒng)、模糊離散事件系統(tǒng)等。賦時(shí)離散事件系統(tǒng)是高時(shí)效性類系統(tǒng),系統(tǒng)的建模不僅需要考慮其運(yùn)行過程的邏輯行為,還需要進(jìn)一步捕捉運(yùn)行過程中的一些時(shí)間信息。它在計(jì)算機(jī)通訊、工業(yè)加工及自動(dòng)化控制等領(lǐng)域都有著極其廣泛的應(yīng)用。故為這類系統(tǒng)尋求一種更安全的故障診斷方法不僅具有理論意義而且具有現(xiàn)實(shí)意義。本文在賦時(shí)離散事件系統(tǒng)的框架下以Paoli等人提出的離散事件系統(tǒng)安全診斷方法為算法基礎(chǔ),針對(duì)一類以自動(dòng)機(jī)為模型的賦時(shí)離散事件系統(tǒng),研究其系統(tǒng)故障診斷的安全性問題,提出一種賦時(shí)離散事件系統(tǒng)的安全診斷方法。首先結(jié)合實(shí)際系統(tǒng)的邏輯信息和時(shí)間信息對(duì)賦時(shí)離散事件進(jìn)行模型的構(gòu)造;然后對(duì)賦時(shí)離散事件系統(tǒng)的安全可診斷性進(jìn)行形式化刻畫;再通過構(gòu)造賦時(shí)離散事件系統(tǒng)非法語言識(shí)別器對(duì)系統(tǒng)的被禁止操作序列進(jìn)行識(shí)別,以保證在故障診斷期間系統(tǒng)若執(zhí)行了被禁止的操作能及時(shí)將其識(shí)別到的信息反饋給系統(tǒng);并在已構(gòu)造的系統(tǒng)非法語言識(shí)別器的基礎(chǔ)上構(gòu)建系統(tǒng)的安全診斷器,用于在線或者離線地診斷系統(tǒng)的故障;最后分析安全診斷器的狀態(tài)得到了一個(gè)關(guān)于賦時(shí)離散事件系統(tǒng)安全可診斷性的充分必要條件,實(shí)現(xiàn)了賦時(shí)離散事件系統(tǒng)的安全診斷,既保證了故障事件一旦發(fā)生之后能及時(shí)被診斷出來,又確保了系統(tǒng)在故障診斷期間不會(huì)執(zhí)行任何不安全操作。最后,給出實(shí)例驗(yàn)證,驗(yàn)證賦時(shí)離散事件系統(tǒng)安全可診斷定理的科學(xué)性。
[Abstract]:Discrete event system is a logical class system which describes the evolution process of the system by state series and event series. The remarkable function of this kind of system is to use control strategy to allow only some policy-compliant event series to occur to meet certain engineering requirements, and is widely used in communication, transportation and other important areas such as public service facilities and computer information processing. With the development of science and technology, the structure of this kind of system is becoming more and more complex, and the fault diagnosis technology in the past is more and more stretched. After that, Sampath et al proposed a diagnostic method based on diagnostics for discrete event systems to solve this problem. In fact, the diagnostic device is constructed based on the diagnostic method of the diagnostics. The diagnostics can run in parallel with the system, and the fault of the system is judged by analyzing the state of the diagnostics. Although the fault diagnosis method of discrete event system proposed by Sampath et al can effectively diagnose system fault, the security problem during system fault diagnosis is not considered. For this reason, Paoli et al further studied the security diagnosis method of discrete event systems on the basis of Sampath. This method has been extended to all kinds of systems to solve the problem of system fault security diagnosis. The systems that have been popularized and applied include stochastic discrete event system, fuzzy discrete event system and so on. Timing discrete event system is a kind of system with high timeliness. The modeling of the system not only needs to consider the logical behavior of its running process, but also needs to further capture some time information in the running process. It is widely used in computer communication, industrial processing and automatic control. Therefore, it is not only of theoretical significance but also practical significance to find a more secure fault diagnosis method for this kind of system. In this paper, based on the security diagnosis method of discrete event systems proposed by Paoli et al., the security of fault diagnosis for a class of time-dependent discrete event systems based on automata is studied in the framework of time-dependent discrete event systems. A security diagnosis method for time-dependent discrete event systems is proposed. Firstly, the model of timing discrete events is constructed according to the logical information and time information of the actual system, and then the security diagnostics of timing discrete event systems are formally characterized. Then the illegal language recognizer of the timing discrete event system is constructed to identify the prohibited operation sequence of the system, so as to ensure that if the prohibited operation is performed during the fault diagnosis, the recognized information can be fed back to the system in time. On the basis of the constructed illegal language recognizer, the security diagnostics of the system are constructed, which can be used to diagnose the faults of the system online or offline. Finally, the state of the security diagnostics is analyzed, and a necessary and sufficient condition for the security diagnostics of the timing discrete event system is obtained, and the security diagnosis of the timing discrete event system is realized. It not only ensures that the fault event can be diagnosed in time after the fault event occurs, but also ensures that the system will not perform any unsafe operation during the fault diagnosis. Finally, an example is given to verify the scientific nature of the security diagnostics theorem for time-dependent discrete event systems.
【學(xué)位授予單位】：廣東工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP277
本文編號(hào)：2482954
[Abstract]:Discrete event system is a logical class system which describes the evolution process of the system by state series and event series. The remarkable function of this kind of system is to use control strategy to allow only some policy-compliant event series to occur to meet certain engineering requirements, and is widely used in communication, transportation and other important areas such as public service facilities and computer information processing. With the development of science and technology, the structure of this kind of system is becoming more and more complex, and the fault diagnosis technology in the past is more and more stretched. After that, Sampath et al proposed a diagnostic method based on diagnostics for discrete event systems to solve this problem. In fact, the diagnostic device is constructed based on the diagnostic method of the diagnostics. The diagnostics can run in parallel with the system, and the fault of the system is judged by analyzing the state of the diagnostics. Although the fault diagnosis method of discrete event system proposed by Sampath et al can effectively diagnose system fault, the security problem during system fault diagnosis is not considered. For this reason, Paoli et al further studied the security diagnosis method of discrete event systems on the basis of Sampath. This method has been extended to all kinds of systems to solve the problem of system fault security diagnosis. The systems that have been popularized and applied include stochastic discrete event system, fuzzy discrete event system and so on. Timing discrete event system is a kind of system with high timeliness. The modeling of the system not only needs to consider the logical behavior of its running process, but also needs to further capture some time information in the running process. It is widely used in computer communication, industrial processing and automatic control. Therefore, it is not only of theoretical significance but also practical significance to find a more secure fault diagnosis method for this kind of system. In this paper, based on the security diagnosis method of discrete event systems proposed by Paoli et al., the security of fault diagnosis for a class of time-dependent discrete event systems based on automata is studied in the framework of time-dependent discrete event systems. A security diagnosis method for time-dependent discrete event systems is proposed. Firstly, the model of timing discrete events is constructed according to the logical information and time information of the actual system, and then the security diagnostics of timing discrete event systems are formally characterized. Then the illegal language recognizer of the timing discrete event system is constructed to identify the prohibited operation sequence of the system, so as to ensure that if the prohibited operation is performed during the fault diagnosis, the recognized information can be fed back to the system in time. On the basis of the constructed illegal language recognizer, the security diagnostics of the system are constructed, which can be used to diagnose the faults of the system online or offline. Finally, the state of the security diagnostics is analyzed, and a necessary and sufficient condition for the security diagnostics of the timing discrete event system is obtained, and the security diagnosis of the timing discrete event system is realized. It not only ensures that the fault event can be diagnosed in time after the fault event occurs, but also ensures that the system will not perform any unsafe operation during the fault diagnosis. Finally, an example is given to verify the scientific nature of the security diagnostics theorem for time-dependent discrete event systems.
【學(xué)位授予單位】：廣東工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP277
【參考文獻(xiàn)】
相關(guān)期刊論文 前5條
1 劉富春;;非確定型離散事件系統(tǒng)雙模擬控制的實(shí)現(xiàn)[J];控制理論與應(yīng)用;2015年01期
2 董煒;陳衛(wèi)征;徐曉濱;吉吟東;;基于可分性測度的模糊隸屬函數(shù)確定方法[J];控制與決策;2014年11期
3 方歡;方賢文;李德權(quán);;基于Petri網(wǎng)的故障診斷研究理論的綜述[J];計(jì)算機(jī)科學(xué);2014年03期
4 王曉宇;歐陽丹彤;趙劍;;不完備模型下的離散事件系統(tǒng)診斷方法[J];軟件學(xué)報(bào);2012年03期
5 宗群;劉利;劉文靜;薛寒冰;;DEDS故障診斷方法在電梯中的應(yīng)用[J];天津大學(xué)學(xué)報(bào);2006年05期
，本文編號(hào)：2482954
本文鏈接：http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2482954.html
最近更新
教材專著
